Product sales potential prediction using Machine Learning models

Adarve Álvarez, Isabel (2022). Product sales potential prediction using Machine Learning models. Thesis (Master thesis), E.T.S. de Ingenieros Informáticos (UPM).

Description

Title: Product sales potential prediction using Machine Learning models
Author/s:
  • Adarve Álvarez, Isabel
Contributor/s:
  • Bajo Pérez, Javier
Item Type: Thesis (Master thesis)
Masters title: Ciencia de Datos
Date: July 2022
Subjects:
Faculty: E.T.S. de Ingenieros Informáticos (UPM)
Department: Inteligencia Artificial
Creative Commons Licenses: Recognition - No derivative works - Non commercial

Full text

[img] PDF - Requires a PDF viewer, such as GSview, Xpdf or Adobe Acrobat Reader
Download (2MB)

Abstract

El presente estudio se enmarca en el ámbito del retail, más concretamente en la predicción del potencial de ventas de un producto. En el contexto de este estudio, a cada producto de mi base de datos se le asocia un valor, llamado potencial de ventas, calculado en base a sus ventas en las tiendas en las que ha estado presente, y teniendo en cuenta los meses en los que dicho producto se enmarca, indicados por su temporada asociada. Para la creación del dataset, hago uso de múltiples features asociadas a los productos, como el color, el precio, la temporada, la composición, y otros atributos extraídos del nombre de los artículos mediante algoritmos de NLP (Natural Language Processing), como puede ser el tipo de mangas o de cuello, el estilo o si tiene o no bolsillos. Todos estos datos han sido extraídos de la base de datos de Nextail Labs S.L.U.[1], y posteriormente anonimizados. Tras la creación de la base de datos a utilizar, hago uso de modelos de Machine Learning para la predicción del potencial de ventas de los productos, analizando los resultados obtenidos a través de una métrica [2] seleccionada para ello.---ABSTRACT---This study is carried out within the framework of fashion retail, more specifically in the prediction of the sales potential of fashion products. In the context of this study, each product in my database is associated with a value, called sales potential, which is calculated using its sales in the stores where it has been sold, and taking into account the season the product may be used for, indicated by the season when it was launched. To create the dataset, I make use of multiple features associated with the products, such as colour, price, season, composition and other attributes extracted from the name of the clothes using NLP (Natural Language Processing) algorithms, such as the type of sleeves or collar, the style or whether it has pockets or not. All this data has been extracted from the Nextail Labs S.L.U. [1] database, and subsequently anonymised. After creating the database, I use Machine Learning models to predict the sales potential of the products, analysing the results obtained through a metric [2] selected for it.

More information

Item ID: 71445
DC Identifier: https://oa.upm.es/71445/
OAI Identifier: oai:oa.upm.es:71445
Deposited by: Biblioteca Facultad de Informatica
Deposited on: 29 Jul 2022 09:44
Last Modified: 29 Jul 2022 09:44
  • Logo InvestigaM (UPM)
  • Logo GEOUP4
  • Logo Open Access
  • Open Access
  • Logo Sherpa/Romeo
    Check whether the anglo-saxon journal in which you have published an article allows you to also publish it under open access.
  • Logo Dulcinea
    Check whether the spanish journal in which you have published an article allows you to also publish it under open access.
  • Logo de Recolecta
  • Logo del Observatorio I+D+i UPM
  • Logo de OpenCourseWare UPM